jibjib-model

jibjib-model copied to clipboard

Hello, I try to run the project but I don't know where to get dataset for training it. Where can I get the data? Then I notice, the input wav...

xav12358

- Move from plain convolutional to recurrent convolutional network - One prediction per 10s audio slice - No more majority vote on received output classes - First prototype working -...

obitech

- Use Keras features and/or generators to minimize RAM usage for training - Also see https://github.com/gojibjib/jibjib/issues/13

obitech

Signal to noise ratio for increasing training quality

2

Implementing intelligent Signal2Noise Ratio Algorithm to cut input audio files into meaningful parts and increase training quality

sebibi07